which 4NT convention?

[jerdonald]

BBO forum,

 

My partner and I have been playing 1430 RKCB for some time.

He recently suggested we go to 3014 and I would like to

go back to straight Blackwood to simplify things.

 

I'm sure there will be a lot of opinions here as to

which is best but is there a way to calculate what

bid would be the most useful over many, many hands?

 

jerryd

[manudude03]

Straight Blackwood is pretty much never played by good pairs these days. It leaves you guessing too often. It's not really possible to calculate numbers without any bias. The only "advantages" of straight blackwood is that it's easier to remember and you're less likely to be forced too high. The first advantage is really non-existent if you're already playing RKC as presumably you don't get misunderstandings from using it. As for the other, the problem is more likely to be judgement, for which straight blackwood would remove in some cases, but gives other problems when missing 1 ace (or 1/2 kings for grand). If you are getting too high with RKC, you can either play kickback instead or just learn not to bid 4NT when you can't handle the responses.

 

As for the 3014 vs 1430 debate, it's personal preference really, 1430 has the theoretical advantage that a 1 key card response is more common than a 0 key card response allowing you to bid 5D as a queen ask when hearts are trumps more often. 3014 is a little more intuitive responses wise (for me anyway).

[Cthulhu D]

It's best to work out cases where an option gains or loses - and yeah, I think that makes the tradeoff something like 1430 is superior when hearts are trumps and you need to find out about the queen, and 0314 is better when clubs are trumps, you cannot tolerate an answer of zero key cards from partner, and you have no other way to find things out.

[Zelandakh]

If you can only use one of these then 1430 is the best, followed by 3041 with ordinary Blackwood some way behind. The reason is that the 1 or 4 key card response is considerably more likely than 0 or 3. If you specifically take out very strong hands then this changes somewhat, since 3 key cards starts to become common enough to offset the frequency of 1 key card. Thus for very strong hands 3041 is better. A few pairs use this to play 1430 for most hands but switch to 3041 in specific situations. This is probably optimal but not a very good idea (imho) for non-expert partnerships. Therefore I would suggest going with 1430 unless one of you finds 3041 significantly easier to remember, in which case the difference between them is not enough to offset occasional forgets.

[barmar]

One suggestion in a book on RKCB is to use 3014 when the strong hand is asking (beceause lower responses are more likely), and 1430 when the weak hand is asking. But if the partnership is seriously considering going back to plain Blackwood, this seems like a complication they wouldn't be interested in (when the hands are about equal, you have to remember rules to determine which response structure you're using).

[PhilKing]

I suspect OP would rather play nothing at all rather than TWO forms of RKCB. Next we'll be sending him a copy of Ken's latest book. The man wants to play something simple - I suggest we let him.

[P_Marlowe]

If you want to simplify things 4NT as 3041 works pretty well, even with hearts as trumps,

or when a minor is agreed.

 

In certain situation you will be forced to get the information via Cue Bidding instead of

keycard asking, but it works.

 

Having the option to ask for the King of trumps, and the Queen of trumps is important,

the specific schema in use is not.

 

With kind regards

Marlowe

[barmar]

Having the option to ask for the King of trumps, and the Queen of trumps is important,

the specific schema in use is not.

Indeed. Bridge magazines occasionally poll champion players on the most important conventions, or ask them about it in interviews, and RKCB is frequently cited for precisely this reason (despite the fact that they also report that Blackwood is one of the most overused conventions by average players).

[fromageGB]

Stick with 1430, as it is the best simple option, and you are already used to it. When you feel like moving on to something better, look together at minorwood and kickback.

[gnasher]

Just play whichever method you're most comfortable with. The benefits of one method over another are far less significant than the cost of getting it wrong.

 

If you and your partner can't agree, offer to swap it for something.

[ArtK78]

One suggestion in a book on RKCB is to use 3014 when the strong hand is asking (beceause lower responses are more likely), and 1430 when the weak hand is asking. But if the partnership is seriously considering going back to plain Blackwood, this seems like a complication they wouldn't be interested in (when the hands are about equal, you have to remember rules to determine which response structure you're using).

Funny. I believe that Kantar recommends exactly the opposite. The default is 1430, but in those rare instances where the weak hand bids RKCB 4NT, the responses are 0314.

[barmar]

Funny. I believe that Kantar recommends exactly the opposite. The default is 1430, but in those rare instances where the weak hand bids RKCB 4NT, the responses are 0314.

That's quite possible. I knew he recommended one or the other, but didn't feel like looking up the details, and just tried to figure it out from intuition. Had a 50% chance of guessing right, that's good enough to bid a vul slam.

[Free]

If you don't have strong feelings about which version to play, just play whatever partner wants and make him happy.

[TylerE]

Learn kickback. When you can guarantee to have 4 steps available below 5suit, everything becomes a lot easier.

[fromageGB]

Learn kickback. When you can guarantee to have 4 steps available below 5suit, everything becomes a lot easier.

As you want to simplify things, you can take advantage of the fact that kickback gives you the same amount of room for responses whatever the suit, and have simple 1,2,3,0 steps in reply for the number of aces. The compression (and confusion) of 3014 or 1430 are not needed...

[Vampyr]

Kickback can be complicated, as it takes a lot of discussion as to when it applies. Unless someone knows of or has written an article about kickback, it might not be the best suggestion for an OP who wants to "simplify things".

[blackshoe]

I know of an article about kickback. Does that make Tyler's suggestion okay?

[Fluffy]

1340 3014 kickback... I would advice you to play no blackwood at all for a shor time period, you would see your judgement improved.

[fromageGB]

1340 3014 kickback... I would advice you to play no blackwood at all for a shor time period, you would see your judgement improved.

I take the point, but "knowledge" of what aces and kings partner has sure beats "judgement" of them. You need judgement (and agreed methods) to decide when a slam might be a good idea, but then knowledge takes over.

[nige1]

Gnasher's advice is good.

Arguably, simpler is better. Whatever else you choose, freeze your decision, for as long as possible, to avoid confusion. .IMO kickback is not that difficult provided that you agree a bid is kickback if it can possibly be kickback.. Also, agree 03/14 or 14/30 once and for all. Consider using an end-signal to disambiguate complex auctions -- it may seem sophisticated but it can actually make things simpler and clearer..

[Zelandakh]

IMO kickback is not that difficult provided that you agree a bid is kickback if it can possibly be kickback.

Funnily enough I recommend precisely the opposite of this. If a bid is logically natural then it stays natural. If it is logically not natural - splinter, cue, etc - then it is Kickback and 4NT gets the meaning of the replaced call. I think this represents the minimum change from standard methods and makes it quite easy to see when Kickback is on and when not. To optimise, more complex rules are useful; but just the basic method is beneficial and good enough on all but a tiny fraction of cases.

[campboy]

It's best to work out cases where an option gains or loses - and yeah, I think that makes the tradeoff something like 1430 is superior when hearts are trumps and you need to find out about the queen, and 0314 is better when clubs are trumps, you cannot tolerate an answer of zero key cards from partner, and you have no other way to find things out.

It's worth bearing in mind that 1430 also gains when clubs are trumps if you are confident partner isn't going to have none but need to find out whether he has one or two (or three). In my experience this has been more frequent than the situation when you need to know whether he has one or none.

[nige1]

Funnily enough I recommend precisely the opposite of this. If a bid is logically natural then it stays natural. If it is logically not natural - splinter, cue, etc - then it is Kickback and 4NT gets the meaning of the replaced call. I think this represents the minimum change from standard methods and makes it quite easy to see when Kickback is on and when not. To optimise, more complex rules are useful; but just the basic method is beneficial and good enough on all but a tiny fraction of cases.

Fair enough. The main thing is to have a rule about when such a bid is kickback. But I prefer Zelandakh's phraseology e.g. I like "when a bid could logically be kickback it is".

[Fluffy]

I take the point, but "knowledge" of what aces and kings partner has sure beats "judgement" of them. You need judgement (and agreed methods) to decide when a slam might be a good idea, but then knowledge takes over.

 

Judgement will force you to understand when blackwood is needed and when i is not. Blackwood is the most missused convention, or it was, before kickback was invented.

[blackshoe]

Funnily enough I recommend precisely the opposite of this. If a bid is logically natural then it stays natural. If it is logically not natural - splinter, cue, etc - then it is Kickback and 4NT gets the meaning of the replaced call. I think this represents the minimum change from standard methods and makes it quite easy to see when Kickback is on and when not. To optimise, more complex rules are useful; but just the basic method is beneficial and good enough on all but a tiny fraction of cases.

1♥-(P)-4♠. Is 4♠ Kickback or natural? Assume, for the sake of argument, that Exclusion KCB is not in the picture.